Work

Joint Patient-Level Analysis of Saphenous Vein Graft Occlusion and Bleeding in the TOP-CABG Randomised Trial

Year: 2026
Type: preprint
Abstract: Text Abstract Background Trials after coronary artery bypass grafting commonly report saphenous vein graft occlusion and bleeding separately, obscuring their overlap within participants. Aims To describe the joint patient-level distribution of one-year saphenous vein graft occlusion and clinically relevant bleeding in a randomised trial comparing 12 and 3 months of dual antiplatelet therapy after coronary artery bypass grafting. Methods This secondary analysis aggregated graft observations for each participant with evaluable one-year imaging. Four mutually exclusive categories described neither outcome, graft occlusion without bleeding, bleeding without graft occlusion, and both. Treatment groups were compared using an unadjusted multinomial model. Occlusion burden and analyses addressing missing imaging were examined. Results Among 2,070 participants with evaluable imaging, 71.8% had neither outcome, 17.3% had graft occlusion without bleeding, 8.5% had bleeding without graft occlusion, and 2.3% had both. The distributions were 69.8%, 16.8%, 10.2%, and 3.2% with 12-month dual antiplatelet therapy and 73.9%, 17.8%, 6.8%, and 1.4% with 3-month therapy (global P = 0.00164). Relative to neither outcome, relative risk ratios were 1.00 (95% confidence interval 0.79–1.26), 1.57 (1.14–2.16), and 2.34 (1.26–4.34), respectively. Among 407 participants with any graft occlusion, 287 had one and 120 had multiple occluded grafts. Analyses addressing missing imaging gave similar directional findings. Conclusions Joint patient-level reporting revealed within-participant overlap between graft occlusion and bleeding and added information on graft-occlusion multiplicity to conventional endpoint summaries. Trial registration number: NCT05380063
Language: English
Subfield: Surgery
Field: Medicine
Open Access status: green
OpenAlex ID: W7225888755
Cites: 12
Cited by:
Related to: 10